Introduction

Quantitatively, oil and oil products are the main environmental pollutants in the world. Exploitation of oil deposits is inevitably accompanied by accidents during extraction, transportation, and storage, and these can lead to the pollution of soil and water. Pollutants are naturally removed by evaporation, photochemical and geochemical reactions, and biodegradation (Komukai-Nakamura et al. 1996; Tjessem and Aaberg 1983).

Bioremediation uses microbes or other biological systems to degrade environmental pollutants (Atlas 1991; Atlas and Atlas 1991; Dua et al. 2002; Heipieper 2007; Owsianiak et al. 2009; Song et al. 1990; Wang and Bartha 1990). Rhizoremediation, which can be used at the final stages of polluted soil re-cultivation, is based on the use of microorganisms to utilize environmental pollutants as carbon and energy source and leads to the destruction of organic pollutants by microorganisms associated with plant roots (Gerhardt et al. 2009; Kuiper et al. 2004). The action of good root-colonizing and pollutant-degrading bacterial strains can result in efficient cleaning of polluted soils. It has been reported that root systems of some plants such as tall fescue, sorghum, maize, alfalfa, ryegrass, Bermuda grass, and rice contain highly potent hydrocarbon-degrading microorganisms (Graj et al. 2013; Hong et al. 2011; Maila et al. 2005; Soleimani et al. 2010).

To elucidate the degrading potential of microorganisms in oil-contaminated soil from Kazakhstan, the present study focused on the isolation and characterization of oil-degrading microorganisms from the rhizosphere of alfalfa (Medicago sativa), grass mixture (Festuca rubra, 75 %; Lolium perenne, 20 %; Poa pratensis, 10 %), and rape (Brassica napus) in Kazakhstan and on the structural characterization of a very large number of aliphatic, alicyclic, and aromatic acids produced during the transformation of a wide range of the hydrocarbon components of crude oil. In order to understand the mechanisms by which the microorganisms degrade the oil components, the isolated species have been cultivated in liquid medium in the presence of oil as sole carbon and energy source, and the supernatants were analyzed chromatographically to separate a wide variety of metabolites identified by GC/MS. The oil degradation, the results of the chemical characterization of a very large number of aliphatic, alicyclic, and aromatic acid products, and the potential of oil decomposition by the isolated species are presented in comparison with the results of well-described species isolated from oil-polluted sand samples collected in the Saudi Arabian Desert and from sewage sludge in Germany. These results are discussed in the context of the shoot and root development of plants which were grown on oil-contaminated sand after inoculation of the seeds with these highly potent newly isolated bacteria.

Materials and methods

Media

Nutrient broth II (SIFIN, Berlin) served as cultivation medium.

Mineral salts medium (MSM, pH 6.3) containing 1.5 g L−1 (NH4)2HPO4, 0.7 g L−1 KH2PO4, 0.8 g L−1 MgSO4 7H2O, 0.5 g L−1 NaCl, 0.46 g L−1 K2SO4, and 20 g L−1 agar (Hundt et al. 1998) was used for isolation of microorganisms from plant rhizosphere and rhizoplane and, without agar, in liquid form for degradation experiments.

Isolation of microorganisms from plant rhizosphere and rhizoplane

Different microorganism strains were isolated from rhizosphere and rhizoplane of plants after pot growth experiments. Each pot was filled with 5 kg of lightly contaminated soil from the Uzen deposit, Mangystau region, Kazakhstan. Soil was analyzed for total content of oil and was supplemented with oil (Uzen deposit, Mangystau region, Kazakhstan) to achieve a final concentration of 20.0 g kg−1. The pots were planted with alfalfa (Medicago sativa), grass mixture (Festuca rubra, 75 %; Lolium perenne, 20 %; Poa pratensis, 10 %), and rape (Brassica napus). The plants were grown in a growth room at RT with 14 h of light period and 10 h of dark period. Duration of cultivation was 50 days.

The plants were removed from the pots, and the bulk soil was shaken off the roots. Ten grams of root sample with adhering soil was washed in 90 mL of sterile distilled water by shaking for 30 min at 25 °C and 220 rpm (Upadhyay et al. 2009). After shaking, 1 mL of the suspension was added to a tube with 9 mL sterile distilled water. This dilution step was repeated three times. The last suspension was used for isolation of rhizosphere microorganisms. The root sample taken from the first flask was washed again in 90 mL of sterile distilled water by shaking for 30 min at 25 °C and 220 rpm; this suspension was also used to isolate microorganisms. The microorganisms of these latter samples were designated as being derived from the rhizoplane of plants.

For isolation of hydrocarbon-oxidizing microorganisms, the plating dilution technique was used. The suspensions were plated on MSM medium. Ten milliliters per liter of oil was added as sole source of carbon and energy.

Identification of isolated microorganisms

Cell morphology (shape, size) of the isolates was determined using a phase-contrast microscope (Axiolab, Zeiss). The Gram characteristics of all bacteria were determined using the KOH test (Suslow et al. 1982).

The three isolates were identified first by their whole cellular fatty acid profiles and matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectra. Biomass of strains grown of Tryptic Soy Agar (medium 535; www.dsmz.de) at 28 °C for 2 days was used for extraction of cellular fatty acids. Cellular fatty acids were analyzed by gas chromatography using the Microbial Identification System (MIDI, Sherlock version 6.1; method, TSBA40; gas chromatograph, model 6890 N, Agilent Technologies) following the method of Stead et al. (1992). Biomass for MALDI-TOF mass spectrometry (MS) analysis was collected after cultivating the strains on Caso Agar (medium 220; www.dsmz.de) at 30 °C for 24 h. Whole-cell protein extracts were obtained according to Protocol 3 of Schumann and Maier (2014) and analyzed by using a Microflex L20 mass spectrometer (Bruker Daltonics). The MALDI-TOF mass spectra were identified with the MALDI-Biotyper software (version 3.1, Bruker Daltonics) using the database update V4.0.0.1.

Molecular genetic analysis for the identification of isolated strains was performed by the National Scientific Shared Laboratory of Biotechnology of National Center for Biotechnology, Kazakhstan. The chromosomal DNA of strains was isolated by the Kate Wilson method (Clayton et al. 1995). Identification of bacteria was performed by direct determination of the nucleotide sequence of the 16S recombinant RNA (rRNA) gene fragment, followed by a BLAST search analysis carried out on the EzTaxon-e server (http://www.ezbiocloud.net/eztaxon; Kim et al. 2012).

Amplification of 16S rRNA fragments

The PCR reaction was performed with universal primers 8 F 5′-AgAgTTTgATCCTggCTCAg-806R-3 and the 5′-ggACTACCAgggTATCTAAT for a total volume of 20 μL. The PCR mixture contained 150 ng of DNA, 1 unit of Maxima Hot Start Taq DNA Polymerase (Fermentas), 0.2 mM of each dNTP, 1× PCR buffer (Fermentas), 2.5 mM MgCl2, and 10 pmol of each primer. The PCR amplification program included long denaturation at 95 °C for 7 min; 35 cycles at 95 °C for 15 s; 52 °C for 30 s; 72 °C for 30 s; and final elongation at 72 °C for 7 min. The PCR program was carried out in a thermocycler GeneAmp PCR System 9700 (Applied Biosystems).

Determination of nucleotide sequence

Purification of PCR products from unrelated primers was performed using the Exonuclease I (Fermentas) and alkaline phosphatase (Shrimp Alkaline Phosphatase, Fermentas) (Werle et al. 1994).

Sequencing reactions were performed using BigDye® Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems) according to the manufacturer’s instructions followed by separation of the fragments on an automated Genetic Analyzer 3730 xl DNA Analyzer (Applied Biosystems).

Reference strains

Nocardia cyriacigeorgica SBUG 1472 was originally isolated from oil-polluted sand samples collected in the Saudi Arabian Desert by enrichment cultivation (Nhi-Cong et al. 2010) and deposited in the strain collection of the Department of Biology of the University Greifswald.

Rhodococcus erythropolis SBUG 271 was originally isolated from sewage sludge collected in Germany by enrichment cultivation (Stope et al. 2002) and deposited in the strain collection of the Department of Biology of the University Greifswald.

Evaluation of growth in the presence of crude oil

Cultivation of bacteria was carried out in 250-mL flasks containing 50 mL of MSM (pH 6.3) supplemented with 1.5 mL of oil. Crude oil from the Uzen deposit, Mangystau region, Kazakhstan, diesel fuel, fuel oil, and gasoline were used as the sole carbon and energy sources at 25 °C and 220 rpm. Assays without substrate and with nutrient broth were used as controls. The dry weight of cultures (biomass grams per liter) was determined after 6 and 10 days of growth by filtering the whole cultures through a glass fiber filter (Whatmann®, Dassel, Germany) which was then dried for 2 h at 100 °C.

Evaluation of growth in the presence of aromatic hydrocarbons

Cultivation of bacteria was carried out on biphenyl and naphthalene. Biphenyl dissolved in diethyl ether was added to sterile 500-mL Erlenmeyer flasks to a final concentration of 100 mg mL−1, naphthalene as a diethyl ether solution to a final concentration of 20 mg mL−1. After evaporation of the diethyl ether over 12 h, 100 mL of MSM was added to each flask, and flasks were shaken for 24 h at 30 °C and 180 rpm to achieve saturation of the compound in the liquid medium. A cell suspension was then added until an optical density (A 500) of 0.04 was reached. Cultures were incubated with biphenyl or naphthalene on a rotary shaker at 30 °C and 180 rpm. Flasks with cell suspension in nutrient broth were used as controls. The mass of cells was determined by the dry weight method for inhomogeneous cell suspensions and by the optical density method for homogeneous cell suspensions.

Degradation experiments

Cultures were shaken in 500-mL Erlenmeyer flasks with 100 mL MSM medium and 1 mL crude oil (Uzen deposit, Mangystau region, Kazakhstan) at 30 °C and 180 rpm. Assays without oil substrate or without cells or with cells and 1 % glucose as sole source of carbon and energy were used as controls.

After 6 days of incubation, the reaction was stopped, the pH value was measured with pH paper, and 5 μL of phenyloctane was added as internal standard. The whole cultures of such transformation assays were first adjusted to pH 9.0 and extracted three times with half a volume diethyl ether. The extracted supernatant was then adjusted to pH 2.0 and extracted further three times with half a volume diethyl ether. The organic phases were dried over anhydrous sodium sulfate and concentrated by rotary evaporation. The residues obtained were dried under a gentle nitrogen stream and dissolved in hexane before analysis. The data are reported as means for two separate experiments with replicated batch cultures. The standard deviation in these replicates was no more than 10 %.

Chemical analysis and identification of products

Gas chromatography–mass spectrometry (GC–MS)

The oil components and metabolites were detected and quantified by injecting 1 μL of the extract into an Agilent gas chromatograph 7890A GC System (Waldbronn, Germany) equipped with a capillary column (Agilent 1901 S-433, 30 m × 250 μm × 0.25 μm, HP-5 ms column) and a mass selective detector 5975C inert XL EI/CI MSD with a quadrupole mass spectrometer. The injector was operated at 250 °C, and the injection was pulsed splitless at 10.5 psi for 2 min. For the analysis of extracts after alkaline extraction, we used elution profile I: the column temperature started at 40 °C and increased to 310 °C at 8 °C min−1 and was finally maintained at 310 °C for 6.25 min. For the analysis of extracts after acidic extraction, we used elution profile II: the column temperature started at 40 °C and increased to 310 °C at 4 °C min−1 and was finally maintained at 310 °C for 7.5 min. To analyze the acid-extractable products, the extracts after acidic extraction were derivatized by methylation with diazomethane as described by De Boer and Backer (1956) in a micro-apparatus (Aldrich-Chemie, Steinheim, Germany). The mass spectrometer conditions were set to 250 °C for the interface; the source conditions were 230 °C; and the quadrupole temperature was set to 150 °C. Analytical standards A and B (A contains C8–C20, 40 mg L−1 each, in hexane; B contains C21–C40, 40 mg L−1 each, in hexane; Sigma-Aldrich, Germany) were used to identify aliphatic oil components.

Microbial inoculation of barley seeds by bacteria

Before inoculation, plant seeds were sterilized with 10 % solution of sodium hypochlorite for 30 min and then washed four times with sterile tap water for 2 h. Seeds were laid out on Luria–Bertani agar medium and incubated for 18–20 h at 24 °C for seed sterility control. Inoculation of the seeds with the isolated bacterial strains was carried out by immersing the seeds into a suspension of bacterial cells (1.3 × 108 cells mL−1) for 2–3 h. The inoculated seeds were sown in plastic vessels filled with 150 g of sterile sand mixed thoroughly with sterile oil at a final concentration of 2 %. Assays without oil substrate or without microbial inoculation and without oil substrate were used as controls. For plant mineral nutrition, Murashige and Skoog basal salt macronutrient solution (Sigma-Aldrich) was used. Plants were grown for 10 days with the following regime: 12 h of light/dark periods at 20 °C. The following model systems were used: barley, barley with oil, and barley with oil and inoculated bacterial strain. To study the effects of inoculation, biometric characteristics of the plants were evaluated by measurement of shoot and root lengths.

Results

We isolated 30 strains of microorganisms from the rhizosphere of plants grown on crude oil as the sole source of carbon and energy. Three of these were selected as very powerful degraders of crude oil components. The L-RP20 strain was isolated from the rhizoplane of Medicago sativa (alfalfa), T-RP18 from the grass mixture (Festuca rubra, 75 %; Lolium perenne, 20 %; Poa pratensis, 10 %), and R-RZ20 from the rhizosphere of Brassica napus (rape).

The Gram-positive isolates were characterized by cellular fatty acid analysis (FAME), MALDI-TOF MS, and 16S rRNA gene fragment analysis. Strain T-RP18 was assigned to the “R. erythropolis group” (consisting of R. erythropolis, Rhodococcus globerulus, Rhodococcus baikonurensis, and Rhodococcus jialingiae; Táncsics et al. 2014) by all three methods. MALDI-TOF MS differentiated strain T-RP18 clearly from R. globerulus (score value only 1.214) but could not differentiate this strain from both R. erythropolis (score value 2.091) and R. baikonurensis (score value 2.143), two species showing a 16S rRNA gene sequence similarity as high as 99.2 %. R. jialingiae (Táncsics et al. 2014) is not yet represented in the MALDI Biotyper database. 16S rRNA gene sequence analysis identified the strains L-RP20 and R-RZ20 as belonging to the genus Gordonia. However, the sequences of the partial 16S rRNA gene (676 and 645 bp, respectively) did not allow an unambiguous differentiation of strains L-RP20 and R-RZ20 at the species level. MALDI-TOF MS is well suited for the identification of taxonomically soundly defined species (Schumann and Maier 2014) and identifies both strains L-RP20 and R-RZ20 as Gordonia rubripertincta with score values >2.0. This identification result is supported by the FAME analyses which revealed similarity indices of 0.805 (strain L-RP20) and 0.751 (strain R-RZ20) to the “Gordonia-rubripertincta-GC subgroup B” as well as by sequence similarities of the partial 16S rRNA gene of 98.1 % (strain L-RP20) and 98.0 % (strain R-RZ20) to the type strain G. rubripertincta NBRC 101908T. The isolated and (tentatively) identified bacteria were deposited at the strain collection of the Department of Biology of the University Greifswald and numbered as G. rubripertincta SBUG 1972 (L-RP20), G. rubripertincta SBUG 1971 (R-RZ20), and R. erythropolis/R. baikonurensis/R. jialingiae SBUG 1968 (T-RP18, further described as Rhodococcus sp. SBUG 1968). The 16S rRNA sequences were deposited in GenBank: G. rubripertincta SBUG 1972 (L-RP20) accession no. KM085968, G. rubripertincta SBUG 1971 (R-RZ20) accession no. KP233861 and Rhodococcus sp. SBUG 1968 (T-RP18) accession no. KP233862.

Growth and crude oil consumption of G. rubripertincta SBUG 1971, G. rubripertincta SBUG 1972, and Rhodococcus sp. SBUG 1968

The dry weight of cultures of G. rubripertincta SBUG 1971, G. rubripertincta SBUG 1972, and Rhodococcus sp. SBUG 1968 increased during cultivation in MSM supplemented with different types of oil as the sole source of carbon and energy. The dry weight of each culture with different oils was determined at 6 and 10 days (Table 1). After 10 days, all species had grown well on every type of oil with the exception of G. rubripertincta SBUG 1972 on fuel oil. However, after 6 days all species showed good growth only on crude oil, while G. rubripertincta SBUG 1972 also grew on diesel fuel and G. rubripertincta SBUG 1971 on fuel oil and gasoline.

Table 1 Growth experiments with the isolated species G. rubripertincta SBUG 1971 (1971), G. rubripertincta SBUG 1972 (1972), and Rhodococcus sp. SBUG 1968 (1968) on 3 % (v/v) of different oils after cultivation for 6 and 10 days and controls
Full size table

To determine the extent of consumption of oil components G. rubripertincta SBUG 1971, G. rubripertincta SBUG 1972, and Rhodococcus sp. SBUG 1968 were incubated in MSM supplemented with 1 mL of crude oil as the sole carbon source. After 6 days of incubation, whole cultures were extracted (after alkalization to pH 9), residues were dissolved in hexane and analyzed by GC/MS [elution profile I] (Supplementary Material Fig. S1).

The volatile n-alkanes nonane, decane, undecane, and in part also dodecane evaporated during the incubation process and the following extraction process nearly completely; therefore, these substances could not be analyzed in the elution profile of extracts after alkaline extraction (pH 9) of MSM medium and crude oil. The other n-alkanes with chain length from C12–C32 were completely degraded by the three strains. The remaining compounds are various branched chain alkanes, aromatics, and heteroaromatic compounds. These results showed that G. rubripertincta SBUG 1971 and SBUG 1972 and Rhodococcus sp. SBUG 1968 consumed 100 % of alkanes with chain length from C12–C32.

Beside the 24 detected n-alkanes 16 n-alkylcyclohexanes with n-alkyl chain length from C2 to C18 were also shown to be present in the crude oil (Table 2). As described above, for the volatile n-alkanes nonane, decane, undecane, and dodecane, the volatile n-alkylcyclohexanes ethylcyclohexane, propylcyclohexane, and butylcyclohexane also evaporated during the incubation and subsequent extraction process, so that these substances could not be detected in the elution profile of extracts after alkaline extraction (pH 9) of MSM medium and crude oil. All of the other n-alkylcyclohexanes with chain length from C5 to C18 were completely transformed by the species G. rubripertincta SBUG 1971 and SBUG 1972 and Rhodococcus sp. SBUG 1968. In addition, all alkylcyclopentanes could be transformed by these species.

Table 2 Components of crude oil detected by GC/MS and after growth of Gordonia rubripertincta SBUG 1971, Gordonia rubripertincta SBUG 1972, and Rhodococcus sp. SBUG 1968 on crude oil
Full size table

Furthermore, various n-alkyl-methylcyclohexanes, monomethyl- and monoethyl-substituted alkanes, branched chain alkanes, monoaromatics, naphthalenes, and biphenyls were detected in the crude oil—some of these were in low quantity but were clearly identifiable (Table 2). The n-alkyl-methylcyclohexanes and methylcyclopentanes, the monomethyl- and monoethyl-substituted alkanes, the alkylmonoaromatics and diaromatics, and the alkylnaphthalenes could not be detected after incubation with the strains G. rubripertincta SBUG 1971 and SBUG 1972 and Rhodococcus sp. SBUG 1968, whereas the bulk of the branched chain alkanes as well as aromatics and heteroaromatics without alkyl chains remained in the medium.

Analyses of aliphatic, alicyclic, and aromatic acids derived from crude oil components

To characterize the degradation pathways for crude oil components, G. rubripertincta SBUG 1971 and SBUG 1972 and Rhodococcus sp. SBUG 1968 were incubated in MSM supplemented with 1 mL of crude oil as the sole carbon source. After 6 days of incubation, whole cultures were extracted after acidification to pH 2, and residues were dissolved in hexane, methylated by diazomethane and analyzed by GC/MS [elution profile II] (Supplementary Material Fig. S2).

During the incubation of G. rubripertincta SBUG 1971 and SBUG 1972 and Rhodococcus sp. SBUG 1968 with crude oil, a large number of acidic products were identified by GC/MS analysis by comparison with standard compounds or with the spectral library of the National Institute of Standard Technology (NIST08). In the control extract of the assay with glucose (1 %) as sole source of carbon and in the control extract of the assay with crude oil without cells, no acidic products were detected. In summary, 44 different acidic products were identified by comparison to reference MS spectra of the NIST08 library or to the retention times and mass spectra of standard compounds (overview Table 3; mass spectrum data Supplementary Material Tables S1, S2, S3, S4, S5, and S6). Only one n-alkanoic acid was detected in low amount and was identified as heptanoic acid (M1), whereas four monoethyl-substituted and two methyl-substituted alkanoic acids (M2–M7) were obtained. Besides the n-alkyl and branched chain alkyl carboxylic acids, seven cyclohexylalkanoic acids (M8–M14) and five cyclopentylalkanoic acids (M15–M19) were identified with cyclohexylacetic acid (M12) being present in the highest yield. The degradation of n-alkylbenzenes led to the formation of phenylpropionic acid (M20), benzoic acid (M21), and 2-hydroxybenzoic acid (M22) or 3,4-dihydroxybenzoic acid (M23) if odd-numbered n-alkylbenzenes were consumed and to phenylacetic acid (M24) in the case of even-numbered substrates. Furthermore, six different alkyl-substituted benzoic acids (M25–M30) were recovered from poly-n-alkyl-substituted benzenes. Additionally, five branched-chain-alkyl-substituted aromatic acids (M31–M35) were identified as transformation products. The transformation of n-alkylnaphthalenes led to the formation of 1-naphthylpropionic acid (M36) and 1-naphthalene-carboxylic acid (M37) from odd-numbered n-alkylnaphthalenes and to 1-naphthylbutyric acid (M38) and 1-naphthylacetic acid (M39) from even-numbered substrates. Two other substituted naphthylacetic (M40) and naphthalene-carboxylic (M41) acids were identified as transformation products. Similar transformations of alkylbiphenyls resulted in the biphenylalkanoic acids M42, M43, and M44.

Table 3 Aliphatic oil components detected by GC/MS and aliphatic acids formed during growth on crude oil by Gordonia rubripertincta SBUG 1971, Gordonia rubripertincta SBUG 1972 and Rhodococcus sp. SBUG 1968
Full size table

Crude oil consumption of reference strains

In summary, G. rubripertincta SBUG 1971 and SBUG 1972 and Rhodococcus sp. SBUG 1968 were able to degrade or transform more than 80 different components of the crude oil. All these components contain at least one alkyl chain and were therefore transformed by alkane monooxygenases or by the very similar class of monooxygenases oxidizing alkane substituents of cycloaliphatic or aromatic rings. However, gordoniae and rhodococci are also able to directly oxidize aromatic rings either by ring-monooxygenases or ring-dioxygenases to form the corresponding epoxides or dihydrodiols and that are transformed to phenolic and catecholic compounds that are then further degraded (Andreoni et al. 2000; Arenskötter et al. 2004; de Carvalho and da Fonseca 2005b; Gallego et al. 2014; Kästner et al. 1994; Kim et al. 2002; Lin et al. 2012; Warhurst and Fewson 1994). The R. erythropolis SBUG 271 is able to hydroxylate and cleave the aromatic ring of biphenyl and dibenzofuran (Stope et al. 2002) and to hydroxylate and further transform carbazole, dibenzothiophene, and fluorene (Waldau et al. 2009). We therefore used R. erythropolis SBUG 271 as reference strain for direct attack at aromatic rings. This strain was incubated with crude oil, and whole cultures were extracted and analyzed as described for the isolated species. All the compounds named in Table 2 were handled by R. erythropolis SBUG 271 in the same way as described for the newly isolated strains with one exception. R. erythropolis SBUG 271, a biphenyl degrader, was able to consume biphenyl in crude oil completely (Supplementary Table S7 A) whereas naphthalene and phenanthrene remained untouched.

Furthermore, G. rubripertincta SBUG 1971, G. rubripertincta SBUG 1972, Rhodococcus sp. SBUG 1968, and also R. erythropolis SBUG 271 were not able to transform polymethyl-substituted alkanes, e.g., 2,6,10-trimethyldodecane, 2,6,10-trimethylpentadecane, pristane (2,6,10,14-tetramethylpentadecane), and 2,6,10,14-tetramethylhexadecane during growth on crude oil, whereas a Rhodococcus sp. strain (Komukai-Nakamura et al. 1996), later identified as R. erythropolis, was reported to degrade pristane. In order to determine if the pristane and other polymethyl-substituted alkanes of the Kazakh crude oil can in principle be degraded by pristane-degrading microorganisms and to compare these results with the results of the Kazakh strains, we used N. cyriacigeorgica SBUG 1472, a pristane degrader (Nhi-Cong et al. 2010), as reference strain. This strain was also incubated, and whole cultures were extracted and analyzed as described for the isolated species. After 7 days of incubation with N. cyriacigeorgica SBUG 1472, the n-alkanes dodecane, tridecane, and tetradecane were completely degraded whereas n-alkanes with progressively longer chain lengths were progressively less degraded (Table 4). The polymethyl-substituted alkanes 2,6,10-trimethyldodecane, pristane (2,6,10,14-tetramethylpentadecane), and 2,6,10,14-tetramethylhexadecane were not transformed, whereas 2,6,10-trimethylpentadecane was degraded to a low degree. After 14 days, all n-alkanes were completely consumed by N. cyriacigeorgica and all the above named polymethyl-substituted alkanes were also utilized to a considerable extent (20–60 %).

Table 4 Degradation of n-alkanes and branched chain alkanes during growth of Nocardia cyriacigeorgica SBUG 1472 on crude oil
Full size table

Growth experiments on aromatic hydrocarbons

Growth of all bacteria was carried out on biphenyl and naphthalene in order to determine if the strains are able to degrade these substrates as pure compounds and to compare these results with their capacity to degrade biphenyl and naphthalene mixed into crude oil (Supplementary Material Table S7). None of the strains could grow on naphthalene, and only R. erythropolis SBUG 271 could use biphenyl as growth substrate. Furthermore, naphthalene in crude oil could not be transformed or degraded and biphenyl in crude oil could only be consumed by R. erythropolis SBUG 271.

Influence of microbial inoculation of barley seeds on the plant development

Barley seeds were inoculated with the isolated bacterial strains G. rubripertincta SBUG 1971, G. rubripertincta SBUG 1972, and Rhodococcus sp. SBUG 1968 and sown in oil-containing sand. Plants were grown for 10 days and then the lengths of shoots and roots were measured (Fig. 1). The best protective effect on plants was observed in systems inoculated with G. rubripertincta SBUG 1971 and SBUG 1972. This treatment of seeds stimulated the growth of barley seedlings, increasing the growth of shoots by 86 and 42 % and of roots by 42 and 78 %, respectively, compared with their growth in oil-containing sand without inoculation. In the case of Rhodococcus sp. SBUG 1968, weak effects on the root system were noted, but the growth of shoots was increased by 100 %. In parallel, experiments with inoculation of barley seeds with bacteria and their growth in sand without oil were conducted. The inoculation of seeds with bacteria had no significant effect on growth of barley. Only inoculation by Rhodococcus sp. SBUG 1968 marginally increased the length of shoot and root.

Fig. 1
figure 1

Influence of microbial inoculation of barley seeds on the plant development in oil-containing sand. G. rubripertincta SBUG 1971 (SBUG 1971), G. rubripertincta SBUG 1972 (SBUG 1972), and Rhodococcus sp. SBUG 1968 (SBUG 1968)

Full size image

Discussion

Three oil-degrading bacterial strains have been isolated from the rhizosphere and characterized in respect to their ability to use different types of oil as the sole carbon and energy source. These microorganisms were identified as G. rubripertincta SBUG 1971, G. rubripertincta SBUG 1972, and Rhodococcus sp. SBUG 1968. Few R. erythropolis strains were also isolated from the rhizosphere (Cirou et al. 2007; Jussila et al. 2006). Rhodococci have been described as major components of the bacterial population of the microenvironment of potato rhizosphere (Diallo et al. 2011). In contrast, the isolation of gordoniae from the rhizosphere is only known from mangrove rhizosphere (Arenskötter et al. 2004; Takeuchi and Hatano 1998). Hydrocarbon utilizing gordoniae have also been isolated from oil-contaminated soils (Kummer et al. 1999; Nicdao and Rivera 2012). Altogether, gordoniae have been found relatively seldom in the rhizosphere but often in bulk soil; therefore, it is unclear if our isolated species can unambiguously be identified as rhizosphere microorganisms.

To study the spectrum of hydrocarbons utilized by our isolated strains, we used low viscosity crude oil with a low content of resins and asphaltenes, and all three strains showed good to excellent growth on the sixth and on the tenth day of incubation. However, diesel fuel, fuel oil, and gasoline have a more viscous structure due to their higher content of sulfur, resins, asphaltenes, and other components and are consequently more difficult to degrade (Gailiūtė et al. 2011; Khorasani et al. 2013). In line with this, Rhodococcus sp. SBUG 1968 grew poorly on medium with diesel fuel, fuel oil, or gasoline by the sixth day of cultivation and G. rubripertincta SBUG 1972 grew poorly on medium with fuel oil by the sixth day. Hydrocarbons utilizing gordoniae and rhodococci have been previously isolated from oil-contaminated sludge and soil (Kummer et al. 1999; Nicdao and Rivera 2012; Song et al. 2011; Young et al. 2005). They have been reported to be able to grow on diesel-containing medium or on a wide range of saturated hydrocarbons and/or polyaromatics.

We show that our isolates can biodegrade an unusually wide spectrum of hydrocarbons as substrates (more than 80 oil components; Table 2). The first important result to be noticed is that all three isolates can transform mainly alkyl chains of components of the crude oil. n-Alkanes with chains ranging from 12 to 32 carbon atoms and monomethyl- and monoethyl-substituted alkanes (C12–C23) were efficiently used as substrates. Gordoniae have been reported to degrade alkanes in the range of C12 to C36 (Kummer et al. 1999; Lo Piccolo et al. 2011; Nicdao and Rivera 2012) and rhodococci can utilize a broad range of branched and n-alkanes with chain length of C5 to C35 (de Carvalho and da Fonseca 2005a; de Carvalho et al. 2005; de Carvalho et al. 2009; Komukai-Nakamura et al. 1996; Rapp and Gabriel-Jürgens 2003; Whyte et al. 1998). The monoterminal degradation pathway for n-alkanes and branched chain alkanes involves primary hydroxylation at the terminal methyl group by alkane 1-monooxygenases (Shen et al. 2010), further oxidation to monocarboxylic acids and ß-oxidation forming various alkanoic acids with different chain length as intermediates (McKenna and Kallio 1971; Rojo 2009; Singh et al. 2012; van Beilen et al. 2003; Watkinson and Morgan 1990; Widdel and Musat 2010). This explains the recovery of n-heptanoic acid (M1), four monoethyl-substituted and two methyl-substituted alkanoic acids (M2–M7) from n-alkane and methyl- and ethyl-substituted alkane transformation by the new isolates (Table 3). The accumulated intermediate n-heptanoic acid (M1) is only one of the intermediates on the pathway to acetyl CoA and propionyl CoA (Fig. 2a) starting from n-alkanes as substrates.

Fig. 2
figure 2

Transformation pathways a of n-alkanes; b of n-alkyl-substituted cyclohexanes; and c of n-alkyl-substituted benzenes, naphthalenes, and biphenyls with identified metabolites (M1–M44)

Full size image

Using alkylcyclohexanes and alkylcyclopentanes and alkyl-substituted benzenes, naphthalenes and biphenyls of the Kazakh crude oil as substrates 7 cyclohexylalkanoic and 5 cyclopentylalkanoic acids as well as 16 phenyl, 6 naphthyl, and 3 biphenyl alkanoic acids were identified as metabolites during transformation by the three isolates (Table 3; Fig. 2b, c). The most common pathway of oxidation of alkyl chains is the monoterminal oxidation via an alcohol and an aldehyde intermediate (Morgan and Watkinson 1994; Ratledge 1978) to the corresponding carboxylic acid. This pathway of microbial transformation has been found for alkyl substituents of cyclohexane (Beam and Perry 1974; Feinberg et al. 1980; Perry 1977; Solano-Serena et al. 2008), heteroaromatics, and aromatics (Awe et al. 2008; Fedorak et al. 1996; Herter et al. 2012; Mikolasch et al. 2003; Poole and Whitaker 1997; Webley et al. 1956; White and Russell 1994). Both carboxylic acid and also acetic acid derivatives were identified from even-numbered and also from odd-numbered n-alkyl substituents (Awe et al. 2008; Beam and Perry 1974; Dutta and Harayama 2001; Fedorak et al. 1988; Fedorak and Westlake 1986; Herter et al. 2012; Koma et al. 2005; Rontani and Bonin 1992; Sariasla et al. 1974). Therefore, it is not entirely clear, whether the carboxylic and acetic acid derivatives derived from even- or from odd-numbered n-alkyl substituted parent compounds if mixtures like crude oil are used as substrates.

On the other hand, degradation of alkylbenzenes by bacteria can also start with an attack on the aromatic ring by oxygenation or simultaneously on both the alkyl side chain by monooxygenation and on the aromatic ring by monooxygenation or dioxygenation (Awe et al. 2009; Bhatia and Singh 1996; Jigami et al. 1979; Smith and Ratledge 1989). We found two ring attacked products—the 2-hydroxybenzoic acid (M22) and the 3,4-dihydroxybenzoic acid (M23)—in the incubation experiments. These are two important intermediates on the pathway of complete degradation of n-alkyl substituted benzenes, but we had no evidence of direct oxidation of aromatic rings either by ring-monooxygenases or ring-dioxygenases before alkyl chain metabolism. Furthermore, naphthalene, biphenyl, and phenanthrene of crude oil were not degraded by these strains, although the degradation of aromatic compounds by gordoniae and rhodococci is well established (Andreoni et al. 2000; Arenskötter et al. 2004; de Carvalho and da Fonseca 2005b; Gallego et al. 2014; Kästner et al. 1994; Kim et al. 2002; Lin et al. 2012; Warhurst and Fewson 1994). R. erythropolis SBUG 271 was able to degrade biphenyl of the Kazakh crude oil as expected for a well-known biphenyl degrader (Stope et al. 2002; Waldau et al. 2009). However, naphthalene and phenanthrene were not transformed in the crude oil or when supplied as a pure substrate, although this strain is described to transform the aromatic rings of dibenzofuran (Stope et al. 2002), carbazole, dibenzothiophene, and fluorene (Waldau et al. 2009). All these aromatics were degraded via cometabolic biotransformation or after preculture with biphenyl and that may explain why they were not degraded in complex mixtures like crude oil during growth without preculture. Our new isolates were not able to grow on biphenyl and naphthalene or to transform these substrates in crude oil. This suggests that these strains might be used in mixed populations with, e.g., biphenyl and polyaromatic hydrocarbon degraders for further experiments.

Cyclohexylacetic acid (M12) was formed by G. rubripertincta SBUG 1971 and SBUG 1972 in higher yield than any of the other acids. Feinberg et al. (1980) reported the further degradation of cyclohexylacetic acid by use of a mixed culture of Mycobacterium and Arthrobacter species indicating the potential of mixed populations for bioremediation of special oil components. The cyclohexane-carboxylic acid (M9) and cyclohexylacetic acid (M12) can be further transformed via different pathways. On one hand side, M9 is metabolized to 1-cyclohexene-1-carboxylic acid (M10) and pimelic acid by continued oxidation (Blakley and Papish 1982; Rontani and Bonin 1992; Solano-Serena et al. 2008). The transformation of M9 to 4-hydroxybenzoic acid via 4-oxocyclohexanecarboxylic acid has been described in some bacteria (Blakley and Papish 1982; Smith and Callely 1975). In contrast, the formation of benzoic acid (M21) from M9 via 1-cyclohexene-1-carboxylic acid (M10) was shown by Dutta and Harayama (2001) and Koma et al. (2005). We have shown the occurrence of M9, M10, and 1,4-cyclohexadiene-1-carboxylic acid (M11) and also benzoic acid (M21), which can also be obtained from the degradation of alkyl-substituted benzenes (Fig. 2b, c). This indicates that the transformation of different components of crude oil is based on convergent degradation pathways. To the best of our knowledge, this is the first report of 1,4-cyclohexadiene-1-carboxylic acid (M11) as a metabolite in the degradation pathway of odd-numbered alkyl-substituted cycloalkane oil components. Although we did not detect any n-alkylcyclohexenes in the crude oil, it is possible that some of these substances are present in traces so that 1-cyclohexene-1-carboxylic acid (M10) and 1,4-cyclohexadiene-1-carboxylic acid (M11) could also be derived from such substances. However, in that case, we should have also detected 1-cyclohexene-1-acetic and 1,4-cyclohexadiene-1-acetic acid, but neither of these compounds were formed.

n-Alkyl-methylcyclohexanes and methylcyclopentanes have been transformed by our new isolates to the products 4-methylcyclohexane-carboxylic acid (M13), 1-cyclohexene-1,2-dicarboxylic acid (M14), 3-methylcyclopentane-carboxylic acid (M17), 3-methyl-2-cyclopentene-1,2-dicarboxylic acid (M18), and 3-methylenecyclopentane-1,2-dicarboxylic acid (M19). This is the first report of the production of 1-cycloalkene-1,2-dicarboxylic acids as metabolites. These cycloalkene-1,2-dicarboxylic acids can be formed from poly-n-alkylcycloalkanes by hydroxylation and further oxidation of two n-alkyl chains in the same manner as described above.

All of the four monoethyl-substituted and two methyl-substituted alkanoic acids (M2–M7) contain the branching point in the α-position to the carboxylic group. From these results—together with the persistence of the polyalkyl-substituted alkanes 2,6,10-trimethyldodecane, 2,6,10-trimethylpentadecane, pristane (2,6,10,14-tetramethylpentadecane), and 2,6,10,14-tetramethylhexadecane—we conclude that our microorganisms are not able to degrade branched chain alkanes completely. They can hydroxylate the terminal methyl group, oxidize the resulting alcohol and the aldehyde, and ß-oxidize the resulting acid up to the branching point. Only a few bacteria such as Rhodococcus ruber, Mycobacterium neoaurum, and N. cyriacigeorgica are able to mineralize and utilize branched chain hydrocarbons completely (Nhi-Cong et al. 2009; Nhi-Cong et al. 2010). As expected, N. cyriacigeorgica SBUG 1472 degraded the polymethyl-substituted alkanes of the Kazakh crude oil, but not before all n-alkanes have been completely consumed (Table 4). This suggests that the degradation of branched chain alkanes is repressed by the consumption of n-alkanes by N. cyriacigeorgica SBUG 1472. Furthermore, it indicates that our newly isolated strains and R. erythropolis SBUG 271 are not able to degrade polymethyl-substituted alkanes because these compounds remained untouched after the total degradation of n-alkanes. Thus, the newly isolated strains should also be used in mixed populations with branched chain alkane degraders for further experiments.

On the basis of all transformation results, we conclude that the main components of the Kazakh crude oil can be degraded by G. rubripertincta SBUG 1971, G. rubripertincta SBUG 1972, and Rhodococcus sp. SBUG 1968. Furthermore, these bacteria isolated from the rhizosphere helped to improve the growth of barley seeds in the presence of crude oil due to detoxification of oil constituents. For rhizoremediation, an important contribution to the degradation of pollutants is attributed to microorganisms (Graj et al. 2013; Kuiper et al. 2004), suggesting that the newly isolated species might be used for rhizoremediation projects in Kazakhstan and elsewhere.